U.S. patent application number 16/689982 was filed with the patent office on 2021-05-20 for way to generate tight 2d bounding boxes for autonomous driving labeling.
The applicant listed for this patent is Baidu USA LLC. Invention is credited to JAEWON JUNG, GUODONG RONG, PEITAO ZHAO.
Application Number | 20210150226 16/689982 |
Document ID | / |
Family ID | 1000004496545 |
Filed Date | 2021-05-20 |
![](/patent/app/20210150226/US20210150226A1-20210520-D00000.png)
![](/patent/app/20210150226/US20210150226A1-20210520-D00001.png)
![](/patent/app/20210150226/US20210150226A1-20210520-D00002.png)
![](/patent/app/20210150226/US20210150226A1-20210520-D00003.png)
![](/patent/app/20210150226/US20210150226A1-20210520-D00004.png)
![](/patent/app/20210150226/US20210150226A1-20210520-D00005.png)
![](/patent/app/20210150226/US20210150226A1-20210520-D00006.png)
![](/patent/app/20210150226/US20210150226A1-20210520-D00007.png)
![](/patent/app/20210150226/US20210150226A1-20210520-D00008.png)
![](/patent/app/20210150226/US20210150226A1-20210520-D00009.png)
![](/patent/app/20210150226/US20210150226A1-20210520-D00010.png)
View All Diagrams
United States Patent
Application |
20210150226 |
Kind Code |
A1 |
RONG; GUODONG ; et
al. |
May 20, 2021 |
WAY TO GENERATE TIGHT 2D BOUNDING BOXES FOR AUTONOMOUS DRIVING
LABELING
Abstract
A method, apparatus, and system for generating tight
two-dimensional (2D) bounding boxes for visible objects in a
three-dimensional (3D) scene is disclosed. A two-dimensional (2D)
segmentation image of a three-dimensional (3D) scene comprising one
or more objects is generated by rendering the 3D scene with a
segmentation camera. Each of the objects is rendered in a single
respective different color. Next, one or more visible objects in
the 3D scene are identified among the one or more objects based on
the segmentation image. Next, a 2D amodal segmentation image for
each of the visible objects in the 3D scene is generated
separately. Each amodal segmentation image comprises only the
single visible object for which it is generated. Thereafter, a 2D
bounding box is generated for each of the visible objects in the 3D
scene based on the amodal segmentation image for the visible
object.
Inventors: |
RONG; GUODONG; (Sunnyvale,
CA) ; ZHAO; PEITAO; (Sunnyvale, CA) ; JUNG;
JAEWON; (Sunnyvale, CA) |
|
Applicant: |
Name |
City |
State |
Country |
Type |
Baidu USA LLC |
Sunnyvale |
CA |
US |
|
|
Family ID: |
1000004496545 |
Appl. No.: |
16/689982 |
Filed: |
November 20, 2019 |
Current U.S.
Class: |
1/1 |
Current CPC
Class: |
G06K 9/6256 20130101;
G06T 2207/20081 20130101; G06T 2207/30261 20130101; G06K 9/00805
20130101; G06T 2207/10024 20130101; G06T 7/90 20170101; G06T 7/11
20170101 |
International
Class: |
G06K 9/00 20060101
G06K009/00; G06T 7/11 20060101 G06T007/11; G06T 7/90 20060101
G06T007/90; G06K 9/62 20060101 G06K009/62 |
Claims
1. A computer-implemented method for perceiving obstacles of
autonomous driving, the method comprising: generating a
two-dimensional (2D) segmentation image of a three-dimensional (3D)
scene comprising one or more objects by rendering the 3D scene with
a segmentation camera, wherein each of the objects is rendered in a
single respective different color; identifying one or more visible
objects in the 3D scene among the one or more objects based on the
segmentation image; generating a 2D amodal segmentation image for
each of the visible objects in the 3D scene separately, wherein
each amodal segmentation image comprises only the single visible
object for which it is generated; and generating a 2D bounding box
for each of the visible objects in the 3D scene based on the amodal
segmentation image for the visible object.
2. The method of claim 1, wherein identifying the one or more
visible objects in the 3D scene based on the segmentation image
further comprises: determining a quantity of visible pixels for
each of the objects in the 3D scene, wherein all pixels present in
the segmentation image associated with an object are visible
pixels, and wherein a correspondence between a pixel and its
associated object is identified based on a color of the pixel and
the single color of the associated object; determining, for each of
the objects, whether a quantity of visible pixels associated with
the object is greater than a predetermined threshold; and
identifying an object in the 3D scene as a visible object in
response to determining that the quantity of visible pixels
associated with the object is greater than the predetermined
threshold.
3. The method of claim 1, wherein generating the 2D bounding box
for each of the visible objects in the 3D scene based on the amodal
segmentation image for the visible object further comprises:
determining a minimum x-coordinate, a minimum y-coordinate, a
maximum x-coordinate, and a maximum y-coordinate associated with
pixels of the visible object based on the amodal segmentation
image; generating a rectangle associated with the minimum
x-coordinate, the minimum y-coordinate, the maximum x-coordinate,
and the maximum y-coordinate as the 2D bounding box for the visible
object.
4. The method of claim 1, wherein each amodal segmentation image
associated with a single visible object is generated with a 1-bit
color depth.
5. The method of claim 4, wherein a plurality of amodal
segmentation images are rendered into a single color-format image
in a single render pass, and wherein each color information bit in
the color-format image corresponds to a respective visible
object.
6. The method of claim 5, wherein 32 amodal segmentation images
associated with 32 visible objects are rendered into a single
32-bit-color-format image in a single render pass.
7. The method of claim 4, wherein a first plurality of amodal
segmentation images are rendered with a multiple render target
(MRT) technique into a second plurality of color-format images in a
single render pass, and wherein each color information bit in each
of the color-format images corresponds to a respective visible
object.
8. The method of claim 7, wherein 128 amodal segmentation images
associated with 128 visible objects are rendered into four
32-bit-color-format images in a single render pass.
9. A non-transitory machine-readable medium having instructions
stored therein, which when executed by a processor, cause the
processor to perform operations for perceiving obstacles of
autonomous driving, the operations comprising: generating a
two-dimensional (2D) segmentation image of a three-dimensional (3D)
scene comprising one or more objects by rendering the 3D scene with
a segmentation camera, wherein each of the objects is rendered in a
single respective different color; identifying one or more visible
objects in the 3D scene among the one or more objects based on the
segmentation image; generating a 2D amodal segmentation image for
each of the visible objects in the 3D scene separately, wherein
each amodal segmentation image comprises only the single visible
object for which it is generated; and generating a 2D bounding box
for each of the visible objects in the 3D scene based on the amodal
segmentation image for the visible object.
10. The non-transitory machine-readable medium of claim 9, wherein
identifying the one or more visible objects in the 3D scene based
on the segmentation image further comprises: determining a quantity
of visible pixels for each of the objects in the 3D scene, wherein
all pixels present in the segmentation image associated with an
object are visible pixels, and wherein a correspondence between a
pixel and its associated object is identified based on a color of
the pixel and the single color of the associated object;
determining, for each of the objects, whether a quantity of visible
pixels associated with the object is greater than a predetermined
threshold; and identifying an object in the 3D scene as a visible
object in response to determining that the quantity of visible
pixels associated with the object is greater than the predetermined
threshold.
11. The non-transitory machine-readable medium of claim 9, wherein
generating the 2D bounding box for each of the visible objects in
the 3D scene based on the amodal segmentation image for the visible
object further comprises: determining a minimum x-coordinate, a
minimum y-coordinate, a maximum x-coordinate, and a maximum
y-coordinate associated with pixels of the visible object based on
the amodal segmentation image; generating a rectangle associated
with the minimum x-coordinate, the minimum y-coordinate, the
maximum x-coordinate, and the maximum y-coordinate as the 2D
bounding box for the visible object.
12. The non-transitory machine-readable medium of claim 9, wherein
each amodal segmentation image associated with a single visible
object is generated with a 1-bit color depth.
13. The non-transitory machine-readable medium of claim 12, wherein
a plurality of amodal segmentation images are rendered into a
single color-format image in a single render pass, and wherein each
color information bit in the color-format image corresponds to a
respective visible object.
14. The non-transitory machine-readable medium of claim 13, wherein
32 amodal segmentation images associated with 32 visible objects
are rendered into a single 32-bit-color-format image in a single
render pass.
15. The non-transitory machine-readable medium of claim 12, wherein
a first plurality of amodal segmentation images are rendered with a
multiple render target (MRT) technique into a second plurality of
color-format images in a single render pass, and wherein each color
information bit in each of the color-format images corresponds to a
respective visible object.
16. The non-transitory machine-readable medium of claim 15, wherein
128 amodal segmentation images associated with 128 visible objects
are rendered into four 32-bit-color-format images in a single
render pass.
17. A data processing system, comprising: a processor; and a memory
coupled to the processor to store instructions, which when executed
by the processor, cause the processor to perform operations for
perceiving obstacles of autonomous driving, the operations
including generating a two-dimensional (2D) segmentation image of a
three-dimensional (3D) scene comprising one or more objects by
rendering the 3D scene with a segmentation camera, wherein each of
the objects is rendered in a single respective different color,
identifying one or more visible objects in the 3D scene among the
one or more objects based on the segmentation image, generating a
2D amodal segmentation image for each of the visible objects in the
3D scene separately, wherein each amodal segmentation image
comprises only the single visible object for which it is generated,
and generating a 2D bounding box for each of the visible objects in
the 3D scene based on the amodal segmentation image for the visible
object.
18. The data processing system of claim 17, wherein identifying the
one or more visible objects in the 3D scene based on the
segmentation image further comprises: determining a quantity of
visible pixels for each of the objects in the 3D scene, wherein all
pixels present in the segmentation image associated with an object
are visible pixels, and wherein a correspondence between a pixel
and its associated object is identified based on a color of the
pixel and the single color of the associated object; determining,
for each of the objects, whether a quantity of visible pixels
associated with the object is greater than a predetermined
threshold; and identifying an object in the 3D scene as a visible
object in response to determining that the quantity of visible
pixels associated with the object is greater than the predetermined
threshold.
19. The data processing system of claim 17, wherein generating the
2D bounding box for each of the visible objects in the 3D scene
based on the amodal segmentation image for the visible object
further comprises: determining a minimum x-coordinate, a minimum
y-coordinate, a maximum x-coordinate, and a maximum y-coordinate
associated with pixels of the visible object based on the amodal
segmentation image; generating a rectangle associated with the
minimum x-coordinate, the minimum y-coordinate, the maximum
x-coordinate, and the maximum y-coordinate as the 2D bounding box
for the visible object.
20. The data processing system of claim 17, wherein each amodal
segmentation image associated with a single visible object is
generated with a 1-bit color depth.
Description
TECHNICAL FIELD
[0001] Embodiments of the present disclosure relate generally to
operating autonomous vehicles. More particularly, embodiments of
the disclosure relate to generating data for training the
perception module for autonomous driving.
BACKGROUND
[0002] Vehicles operating in an autonomous mode (e.g., driverless)
can relieve occupants, especially the driver, from some
driving-related responsibilities. When operating in an autonomous
mode, the vehicle can navigate to various locations using onboard
sensors, allowing the vehicle to travel with minimal human
interaction or in some cases without any passengers.
[0003] Motion planning and control are critical operations in
autonomous driving. However, conventional motion planning
operations estimate the difficulty of completing a given path
mainly from its curvature and speed, without considering the
differences in features for different types of vehicles. Same
motion planning and control is applied to all types of vehicles,
which may not be accurate and smooth under some circumstances.
[0004] The perception module is the key component in the stack of
autonomous driving. Artificial Intelligence (AI) algorithms used in
the perception module require a large amount of labeled images for
training. Manual labeling is both time-consuming and costly, and
can be inaccurate. Accordingly, synthetic data is sometimes used in
the art to generate labeled data to help the perception module
achieve better results.
[0005] Ideally, in a labeled image, each obstacle relevant to
autonomous driving is labeled with a tight two-dimensional (2D)
bounding box. In existing synthetic datasets usable for autonomous
driving, some (e.g., Playing for Data, SYNTHIA) provide no 2D
bounding boxes, while some others (e.g., Ford Center for Autonomous
Vehicles "FCAV," Playing for Benchmarks) have 2D bounding boxes
only for visible pixels (as opposed to 2D bounding boxes for all,
including occluded or truncated, pixels of visible objects), and
still some others (e.g., Virtual KITTI) provide 2D bounding boxes
bigger than the objects themselves that are generated based on
three-dimensional (3D) bounding boxes. To obtain the best training
result, the perception module requires a tight 2D bounding box
which at the same time covers both visible and occluded or
truncated parts of an object.
BRIEF DESCRIPTION OF THE DRAWINGS
[0006] Embodiments of the disclosure are illustrated by way of
example and not limitation in the figures of the accompanying
drawings in which like references indicate similar elements.
[0007] FIG. 1 is a block diagram illustrating a networked system
according to one embodiment.
[0008] FIG. 2 is a block diagram illustrating an example of an
autonomous vehicle according to one embodiment.
[0009] FIGS. 3A-3B are block diagrams illustrating an example of a
perception and planning system used with an autonomous vehicle
according to one embodiment.
[0010] FIG. 4 is a diagram illustrating various types of bounding
boxes described herein.
[0011] FIG. 5 is a block diagram illustrating various components
utilized according to embodiments of the disclosure.
[0012] FIGS. 6A-D are images generated for and illustrative of
embodiments of the disclosure.
[0013] FIG. 7 is a flowchart illustrating an example method for
generating tight two-dimensional (2D) bounding boxes for visible
objects in a three-dimensional (3D) scene according to one
embodiment.
[0014] FIG. 8 is a block diagram illustrating an example apparatus
according to one embodiment.
DETAILED DESCRIPTION
[0015] Various embodiments and aspects of the disclosures will be
described with reference to details discussed below, and the
accompanying drawings will illustrate the various embodiments. The
following description and drawings are illustrative of the
disclosure and are not to be construed as limiting the disclosure.
Numerous specific details are described to provide a thorough
understanding of various embodiments of the present disclosure.
However, in certain instances, well-known or conventional details
are not described in order to provide a concise discussion of
embodiments of the present disclosures.
[0016] Reference in the specification to "one embodiment" or "an
embodiment" means that a particular feature, structure, or
characteristic described in conjunction with the embodiment can be
included in at least one embodiment of the disclosure. The
appearances of the phrase "in one embodiment" in various places in
the specification do not necessarily all refer to the same
embodiment.
[0017] Some embodiments relate to a method, apparatus, and system
for generating tight two-dimensional (2D) bounding boxes for
visible objects in a three-dimensional (3D) scene. First, a
two-dimensional (2D) segmentation image of a three-dimensional (3D)
scene comprising one or more objects is generated by rendering the
3D scene with a segmentation camera. Each of the objects is
rendered in a single respective different color. Next, one or more
visible objects in the 3D scene are identified among the one or
more objects based on the segmentation image. Next, a 2D amodal
segmentation image for each of the visible objects in the 3D scene
is generated separately. Each amodal segmentation image comprises
only the single visible object for which it is generated, and the
whole visible object is rendered in its entirety without occlusion
in the respective amodal segmentation image, even if it would be
partially occluded by another object in the scene when rendered
with a conventional sensor camera. Thereafter, a 2D bounding box is
generated for each of the visible objects in the 3D scene based on
the amodal segmentation image for the visible object.
[0018] The dataset including the generated 2D bounding boxes may be
used to train the perception module of the autonomous vehicle.
[0019] In one embodiment, to identify the one or more visible
objects in the 3D scene based on the segmentation image, a quantity
of visible pixels for each of the objects in the 3D scene is
determined. It should be appreciated that all pixels present in the
segmentation image associated with an object are visible pixels,
and that a correspondence between a pixel and its associated object
can be identified based on a color of the pixel and the single
color of the associated object. Next, for each of the objects,
whether a quantity of visible pixels associated with the object is
greater than a predetermined threshold is determined. Thereafter,
an object in the 3D scene is identified as a visible object if the
quantity of visible pixels associated with the object is greater
than the predetermined threshold.
[0020] In one embodiment, to generate the 2D bounding box for each
of the visible objects in the 3D scene based on the amodal
segmentation image for the visible object, a minimum x-coordinate,
a minimum y-coordinate, a maximum x-coordinate, and a maximum
y-coordinate associated with pixels of the visible object are
determined based on the amodal segmentation image. Then, a
rectangle associated with the minimum x-coordinate, the minimum
y-coordinate, the maximum x-coordinate, and the maximum
y-coordinate is generated as the 2D bounding box for the visible
object.
[0021] In one embodiment, each amodal segmentation image associated
with a single visible object is generated with a 1-bit (per pixel)
color depth. Accordingly, a plurality of amodal segmentation images
can be rendered into a single color-format image in a single render
pass, where each color information bit in the color-format image
corresponds to a respective visible object. For example, in one
embodiment, 32 amodal segmentation images associated with 32
visible objects are rendered into a single 32-bit (per
pixel)-color-format image (e.g., a red green blue alpha "RGBA"
image) in a single render pass.
[0022] In one embodiment, with a multiple render target (MRT)
technique, even more amodal segmentation images can be rendered in
a single render pass. In particular, a first plurality of amodal
segmentation images may be rendered into a second plurality of
color-format images in a single render pass, where each color
information bit in each of the color-format images corresponds to a
respective visible object. For example, in one embodiment, 128
amodal segmentation images associated with 128 visible objects are
rendered into four 32-bit (per pixel)-color-format images (e.g.,
RGBA images) in a single render pass.
[0023] FIG. 1 is a block diagram illustrating an autonomous vehicle
network configuration according to one embodiment of the
disclosure. Referring to FIG. 1, network configuration 100 includes
autonomous vehicle 101 that may be communicatively coupled to one
or more servers 103-104 over a network 102. Although there is one
autonomous vehicle shown, multiple autonomous vehicles can be
coupled to each other and/or coupled to servers 103-104 over
network 102. Network 102 may be any type of networks such as a
local area network (LAN), a wide area network (WAN) such as the
Internet, a cellular network, a satellite network, or a combination
thereof, wired or wireless. Server(s) 103-104 may be any kind of
servers or a cluster of servers, such as Web or cloud servers,
application servers, backend servers, or a combination thereof.
Servers 103-104 may be data analytics servers, content servers,
traffic information servers, map and point of interest (MPOI)
servers, or location servers, etc.
[0024] An autonomous vehicle refers to a vehicle that can be
configured to in an autonomous mode in which the vehicle navigates
through an environment with little or no input from a driver. Such
an autonomous vehicle can include a sensor system having one or
more sensors that are configured to detect information about the
environment in which the vehicle operates. The vehicle and its
associated controller(s) use the detected information to navigate
through the environment. Autonomous vehicle 101 can operate in a
manual mode, a full autonomous mode, or a partial autonomous
mode.
[0025] In one embodiment, autonomous vehicle 101 includes, but is
not limited to, perception and planning system 110, vehicle control
system 111, wireless communication system 112, user interface
system 113, and sensor system 115. Autonomous vehicle 101 may
further include certain common components included in ordinary
vehicles, such as, an engine, wheels, steering wheel, transmission,
etc., which may be controlled by vehicle control system 111 and/or
perception and planning system 110 using a variety of communication
signals and/or commands, such as, for example, acceleration signals
or commands, deceleration signals or commands, steering signals or
commands, braking signals or commands, etc.
[0026] Components 110-115 may be communicatively coupled to each
other via an interconnect, a bus, a network, or a combination
thereof. For example, components 110-115 may be communicatively
coupled to each other via a controller area network (CAN) bus. A
CAN bus is a vehicle bus standard designed to allow
microcontrollers and devices to communicate with each other in
applications without a host computer. It is a message-based
protocol, designed originally for multiplex electrical wiring
within automobiles, but is also used in many other contexts.
[0027] Referring now to FIG. 2, in one embodiment, sensor system
115 includes, but it is not limited to, one or more cameras 211,
global positioning system (GPS) unit 212, inertial measurement unit
(IMU) 213, radar unit 214, and a light detection and range (LIDAR)
unit 215. GPS system 212 may include a transceiver operable to
provide information regarding the position of the autonomous
vehicle. IMU unit 213 may sense position and orientation changes of
the autonomous vehicle based on inertial acceleration. Radar unit
214 may represent a system that utilizes radio signals to sense
objects within the local environment of the autonomous vehicle. In
some embodiments, in addition to sensing objects, radar unit 214
may additionally sense the speed and/or heading of the objects.
LIDAR unit 215 may sense objects in the environment in which the
autonomous vehicle is located using lasers. LIDAR unit 215 could
include one or more laser sources, a laser scanner, and one or more
detectors, among other system components. Cameras 211 may include
one or more devices to capture images of the environment
surrounding the autonomous vehicle. Cameras 211 may be still
cameras and/or video cameras. A camera may be mechanically movable,
for example, by mounting the camera on a rotating and/or tilting a
platform.
[0028] Sensor system 115 may further include other sensors, such
as, a sonar sensor, an infrared sensor, a steering sensor, a
throttle sensor, a braking sensor, and an audio sensor (e.g.,
microphone). An audio sensor may be configured to capture sound
from the environment surrounding the autonomous vehicle. A steering
sensor may be configured to sense the steering angle of a steering
wheel, wheels of the vehicle, or a combination thereof. A throttle
sensor and a braking sensor sense the throttle position and braking
position of the vehicle, respectively. In some situations, a
throttle sensor and a braking sensor may be integrated as an
integrated throttle/braking sensor.
[0029] In one embodiment, vehicle control system 111 includes, but
is not limited to, steering unit 201, throttle unit 202 (also
referred to as an acceleration unit), and braking unit 203.
Steering unit 201 is to adjust the direction or heading of the
vehicle. Throttle unit 202 is to control the speed of the motor or
engine that in turn controls the speed and acceleration of the
vehicle. Braking unit 203 is to decelerate the vehicle by providing
friction to slow the wheels or tires of the vehicle. Note that the
components as shown in FIG. 2 may be implemented in hardware,
software, or a combination thereof.
[0030] Referring back to FIG. 1, wireless communication system 112
is to allow communication between autonomous vehicle 101 and
external systems, such as devices, sensors, other vehicles, etc.
For example, wireless communication system 112 can wirelessly
communicate with one or more devices directly or via a
communication network, such as servers 103-104 over network 102.
Wireless communication system 112 can use any cellular
communication network or a wireless local area network (WLAN),
e.g., using WiFi to communicate with another component or system.
Wireless communication system 112 could communicate directly with a
device (e.g., a mobile device of a passenger, a display device, a
speaker within vehicle 101), for example, using an infrared link,
Bluetooth, etc. User interface system 113 may be part of peripheral
devices implemented within vehicle 101 including, for example, a
keyboard, a touch screen display device, a microphone, and a
speaker, etc.
[0031] Some or all of the functions of autonomous vehicle 101 may
be controlled or managed by perception and planning system 110,
especially when operating in an autonomous driving mode. Perception
and planning system 110 includes the necessary hardware (e.g.,
processor(s), memory, storage) and software (e.g., operating
system, planning and routing programs) to receive information from
sensor system 115, control system 111, wireless communication
system 112, and/or user interface system 113, process the received
information, plan a route or path from a starting point to a
destination point, and then drive vehicle 101 based on the planning
and control information. Alternatively, perception and planning
system 110 may be integrated with vehicle control system 111.
[0032] For example, a user as a passenger may specify a starting
location and a destination of a trip, for example, via a user
interface. Perception and planning system 110 obtains the trip
related data. For example, perception and planning system 110 may
obtain location and route information from an MPOI server, which
may be a part of servers 103-104. The location server provides
location services and the MPOI server provides map services and the
POIs of certain locations. Alternatively, such location and MPOI
information may be cached locally in a persistent storage device of
perception and planning system 110.
[0033] While autonomous vehicle 101 is moving along the route,
perception and planning system 110 may also obtain real-time
traffic information from a traffic information system or server
(TIS). Note that servers 103-104 may be operated by a third party
entity. Alternatively, the functionalities of servers 103-104 may
be integrated with perception and planning system 110. Based on the
real-time traffic information, MPOI information, and location
information, as well as real-time local environment data detected
or sensed by sensor system 115 (e.g., obstacles, objects, nearby
vehicles), perception and planning system 110 can plan an optimal
route and drive vehicle 101, for example, via control system 111,
according to the planned route to reach the specified destination
safely and efficiently.
[0034] Server 103 may be a data analytics system to perform data
analytics services for a variety of clients. In one embodiment,
data analytics system 103 includes data collector 121 and machine
learning engine 122. Data collector 121 collects driving statistics
123 from a variety of vehicles, either autonomous vehicles or
regular vehicles driven by human drivers. Driving statistics 123
include information indicating the driving commands (e.g.,
throttle, brake, steering commands) issued and responses of the
vehicles (e.g., speeds, accelerations, decelerations, directions)
captured by sensors of the vehicles at different points in time.
Driving statistics 123 may further include information describing
the driving environments at different points in time, such as, for
example, routes (including starting and destination locations),
MPOIs, road conditions, weather conditions, etc.
[0035] Based on driving statistics 123, machine learning engine 122
generates or trains a set of rules, algorithms, and/or predictive
models 124 for a variety of purposes. In particular, for example,
the machine learning engine 122 may help improve the perception and
planning of the ADV by generating or training models based on
synthetic data according to embodiments of the disclosure.
Perception training system or module 125 is configured to train a
perception module or model to perform perception of obstacles based
on the images captured by sensors. Algorithms/models 124 can then
be uploaded on ADVs to be utilized during autonomous driving in
real-time.
[0036] FIGS. 3A and 3B are block diagrams illustrating an example
of a perception and planning system used with an autonomous vehicle
according to one embodiment. System 300 may be implemented as a
part of autonomous vehicle 101 of FIG. 1 including, but is not
limited to, perception and planning system 110, control system 111,
and sensor system 115. Referring to FIGS. 3A-3B, perception and
planning system 110 includes, but is not limited to, localization
module 301, perception module 302, prediction module 303, decision
module 304, planning module 305, control module 306, routing module
307.
[0037] Some or all of modules 301-307 may be implemented in
software, hardware, or a combination thereof. For example, these
modules may be installed in persistent storage device 352, loaded
into memory 351, and executed by one or more processors (not
shown). Note that some or all of these modules may be
communicatively coupled to or integrated with some or all modules
of vehicle control system 111 of FIG. 2. Some of modules 301-307
may be integrated together as an integrated module.
[0038] Localization module 301 determines a current location of
autonomous vehicle 300 (e.g., leveraging GPS unit 212) and manages
any data related to a trip or route of a user. Localization module
301 (also referred to as a map and route module) manages any data
related to a trip or route of a user. A user may log in and specify
a starting location and a destination of a trip, for example, via a
user interface. Localization module 301 communicates with other
components of autonomous vehicle 300, such as map and route
information 311, to obtain the trip related data. For example,
localization module 301 may obtain location and route information
from a location server and a map and POI (MPOI) server. A location
server provides location services and an MPOI server provides map
services and the POIs of certain locations, which may be cached as
part of map and route information 311. While autonomous vehicle 300
is moving along the route, localization module 301 may also obtain
real-time traffic information from a traffic information system or
server.
[0039] Based on the sensor data provided by sensor system 115 and
localization information obtained by localization module 301, a
perception of the surrounding environment is determined by
perception module 302. The perception information may represent
what an ordinary driver would perceive surrounding a vehicle in
which the driver is driving. The perception can include the lane
configuration, traffic light signals, a relative position of
another vehicle, a pedestrian, a building, crosswalk, or other
traffic related signs (e.g., stop signs, yield signs), etc., for
example, in a form of an object. The lane configuration includes
information describing a lane or lanes, such as, for example, a
shape of the lane (e.g., straight or curvature), a width of the
lane, how many lanes in a road, one-way or two-way lane, merging or
splitting lanes, exiting lane, etc.
[0040] Perception module 302 may include a computer vision system
or functionalities of a computer vision system to process and
analyze images captured by one or more cameras in order to identify
objects and/or features in the environment of autonomous vehicle.
The objects can include traffic signals, road way boundaries, other
vehicles, pedestrians, and/or obstacles, etc. The computer vision
system may use an object recognition algorithm, video tracking, and
other computer vision techniques. In some embodiments, the computer
vision system can map an environment, track objects, and estimate
the speed of objects, etc. Perception module 302 can also detect
objects based on other sensors data provided by other sensors such
as a radar and/or LIDAR.
[0041] For each of the objects, prediction module 303 predicts what
the object will behave under the circumstances. The prediction is
performed based on the perception data perceiving the driving
environment at the point in time in view of a set of map/rout
information 311 and traffic rules 312. For example, if the object
is a vehicle at an opposing direction and the current driving
environment includes an intersection, prediction module 303 will
predict whether the vehicle will likely move straight forward or
make a turn. If the perception data indicates that the intersection
has no traffic light, prediction module 303 may predict that the
vehicle may have to fully stop prior to enter the intersection. If
the perception data indicates that the vehicle is currently at a
left-turn only lane or a right-turn only lane, prediction module
303 may predict that the vehicle will more likely make a left turn
or right turn respectively.
[0042] For each of the objects, decision module 304 makes a
decision regarding how to handle the object. For example, for a
particular object (e.g., another vehicle in a crossing route) as
well as its metadata describing the object (e.g., a speed,
direction, turning angle), decision module 304 decides how to
encounter the object (e.g., overtake, yield, stop, pass). Decision
module 304 may make such decisions according to a set of rules such
as traffic rules or driving rules 312, which may be stored in
persistent storage device 352.
[0043] Routing module 307 is configured to provide one or more
routes or paths from a starting point to a destination point. For a
given trip from a start location to a destination location, for
example, received from a user, routing module 307 obtains route and
map information 311 and determines all possible routes or paths
from the starting location to reach the destination location.
Routing module 307 may generate a reference line in a form of a
topographic map for each of the routes it determines from the
starting location to reach the destination location. A reference
line refers to an ideal route or path without any interference from
others such as other vehicles, obstacles, or traffic condition.
That is, if there is no other vehicle, pedestrians, or obstacles on
the road, an ADV should exactly or closely follows the reference
line. The topographic maps are then provided to decision module 304
and/or planning module 305. Decision module 304 and/or planning
module 305 examine all of the possible routes to select and modify
one of the most optimal routes in view of other data provided by
other modules such as traffic conditions from localization module
301, driving environment perceived by perception module 302, and
traffic condition predicted by prediction module 303. The actual
path or route for controlling the ADV may be close to or different
from the reference line provided by routing module 307 dependent
upon the specific driving environment at the point in time.
[0044] Based on a decision for each of the objects perceived,
planning module 305 plans a path or route for the autonomous
vehicle, as well as driving parameters (e.g., distance, speed,
and/or turning angle), using a reference line provided by routing
module 307 as a basis. That is, for a given object, decision module
304 decides what to do with the object, while planning module 305
determines how to do it. For example, for a given object, decision
module 304 may decide to pass the object, while planning module 305
may determine whether to pass on the left side or right side of the
object. Planning and control data is generated by planning module
305 including information describing how vehicle 300 would move in
a next moving cycle (e.g., next route/path segment). For example,
the planning and control data may instruct vehicle 300 to move 10
meters at a speed of 30 mile per hour (mph), then change to a right
lane at the speed of 25 mph.
[0045] Based on the planning and control data, control module 306
controls and drives the autonomous vehicle, by sending proper
commands or signals to vehicle control system 111, according to a
route or path defined by the planning and control data. The
planning and control data include sufficient information to drive
the vehicle from a first point to a second point of a route or path
using appropriate vehicle settings or driving parameters (e.g.,
throttle, braking, steering commands) at different points in time
along the path or route.
[0046] In one embodiment, the planning phase is performed in a
number of planning cycles, also referred to as driving cycles, such
as, for example, in every time interval of 100 milliseconds (ms).
For each of the planning cycles or driving cycles, one or more
control commands will be issued based on the planning and control
data. That is, for every 100 ms, planning module 305 plans a next
route segment or path segment, for example, including a target
position and the time required for the ADV to reach the target
position. Alternatively, planning module 305 may further specify
the specific speed, direction, and/or steering angle, etc. In one
embodiment, planning module 305 plans a route segment or path
segment for the next predetermined period of time such as 5
seconds. For each planning cycle, planning module 305 plans a
target position for the current cycle (e.g., next 5 seconds) based
on a target position planned in a previous cycle. Control module
306 then generates one or more control commands (e.g., throttle,
brake, steering control commands) based on the planning and control
data of the current cycle.
[0047] Note that decision module 304 and planning module 305 may be
integrated as an integrated module. Decision module 304/planning
module 305 may include a navigation system or functionalities of a
navigation system to determine a driving path for the autonomous
vehicle. For example, the navigation system may determine a series
of speeds and directional headings to affect movement of the
autonomous vehicle along a path that substantially avoids perceived
obstacles while generally advancing the autonomous vehicle along a
roadway-based path leading to an ultimate destination. The
destination may be set according to user inputs via user interface
system 113. The navigation system may update the driving path
dynamically while the autonomous vehicle is in operation. The
navigation system can incorporate data from a GPS system and one or
more maps so as to determine the driving path for the autonomous
vehicle.
[0048] FIG. 4 shows a process of training a perception module
according to one embodiment. The process can be performed by a
perception training module as shown in FIG. 8, which will be
described in details further below. Referring to FIG. 4, a diagram
400 illustrating various types of bounding boxes described herein
is shown. A bounding box 402 that encloses the vehicle 422 directly
ahead is a tight 2D bounding box. As the vehicle 424 further ahead
is partially occluded by the vehicle 422 in the scene, a 2D
bounding box 404 enclosing only the visible portion of the vehicle
424 is a 2D bounding box for only the visible pixels (such a
bounding box 404 may be referred to as a bounding box with
occlusion). An additional 2D bounding box 406 enclosing all of the
vehicle 424 including the occluded part is also shown. The bounding
box 406 and others like it may be referred to as bounding boxes
without occlusion. A 3D bounding box 408 enclosing the vehicle 426
to the right is generated based on the 3D model of the vehicle 426
(i.e., the 3D bounding box 408 corresponds to minimum and maximum
x-, y-, and z-coordinates associated with the 3D model of the
vehicle 426). A non-tight 2D bounding box 410 can be generated
based on the 3D bounding box 408 by projecting the 3D bounding box
408 to the image plane and generating a 2D bounding box 410 based
on the projected 3D bounding box 408. It should be appreciated that
the non-tight 2D bounding box 410 and others like it are not ideal
for the purpose of training the perception module for autonomous
driving. Finally, a tight 2D bounding box 412 enclosing the vehicle
426 is also shown. Tight 2D bounding boxes without occlusion, such
as bounding boxes 402, 406, and 412 are the most useful for
training the perception module. Thus, various embodiments of the
disclosure are directed to generating tight 2D bounding boxes.
[0049] Referring to FIG. 5, a block diagram 500 illustrating
various components utilized according to embodiments of the
disclosure is shown. Referring further to FIGS. 6A-D, images 506,
510, 514, 600D generated for and illustrative of embodiments of the
disclosure are shown. FIG. 6A illustrates a conventional RGB image
506 generated by rendering a three-dimensional (3D) scene 502 using
a conventional sensor camera 504. A number of objects (obstacles)
503 in the 3D scene 502 are of interest to autonomous driving, such
as pedestrians, automobiles, motorcycles, motorcyclists, bicycles,
and bicycles, etc. FIG. 6B illustrates a two-dimensional (2D)
segmentation image 510. The two-dimensional (2D) segmentation image
510 of the three-dimensional (3D) scene 502 comprising one or more
objects 503 is generated by rendering the 3D scene 502 with a
segmentation camera 508. As shown in FIG. 6B, each of the objects
503 is rendered in a single respective different color. Next, one
or more visible objects in the 3D scene 502 are identified among
the one or more objects 503 based on the segmentation image 510. It
should be appreciated that hereinafter a "camera" in a 3D rendering
context refers to a process that simulates visual phenomena
resulting from optical characteristics of cameras. Some of the
simulated special cameras described herein (e.g., the segmentation
camera 508, the amodal segmentation camera) have imaginary
properties and do not have real world counterparts.
[0050] In one embodiment, to identify the one or more visible
objects in the 3D scene 502 based on the segmentation image 510, a
quantity of visible pixels for each of the objects 503 in the 3D
scene 502 is determined. It should be appreciated that all pixels
present in the segmentation image 510 associated with an object 503
are visible pixels, and that a correspondence between a pixel and
its associated object 503 can be identified based on a color of the
pixel and the single color of the associated object. Next, for each
of the objects 503, whether a quantity of visible pixels associated
with the object 503 is greater than (in a different embodiment, the
relationship may be "equal to or greater than") a predetermined
threshold is determined. Thereafter, an object 503 in the 3D scene
502 is identified as a visible object if the quantity of visible
pixels associated with the object 503 is greater than (in a
different embodiment, the relationship may be "equal to or greater
than") the predetermined threshold.
[0051] Next, a 2D amodal segmentation image 514 for each of the
visible objects in the 3D scene is generated separately. Each
amodal segmentation image 512 comprises only the single visible
object for which it is generated. An amodal segmentation image 514
is generated by rendering the respective object with an amodal
segmentation camera 512. FIG. 6C illustrates the amodal
segmentation images 514 for several objects. Thereafter, a 2D
bounding box 516 is generated for each of the visible objects in
the 3D scene 502 based on the amodal segmentation image 514 for the
visible object.
[0052] In one embodiment, to generate the 2D bounding box 516 for
each of the visible objects in the 3D scene based on the amodal
segmentation image 514 for the visible object, a minimum
x-coordinate, a minimum y-coordinate, a maximum x-coordinate, and a
maximum y-coordinate associated with pixels of the visible object
are determined based on the amodal segmentation image 514. Then, a
rectangle associated with the minimum x-coordinate, the minimum
y-coordinate, the maximum x-coordinate, and the maximum
y-coordinate is generated as the 2D bounding box 516 for the
visible object. FIG. 6D illustrates the result image 600D generated
by superimposing bounding boxes 516 for the visible objects on the
RGB image 506.
[0053] As only the pixel coordinates are useful in an amodal
segmentation image 514, in one embodiment, each amodal segmentation
image 514 associated with a single visible object is generated with
a 1-bit (per pixel) color depth (as is shown in FIG. 6C).
Accordingly, a plurality of amodal segmentation images 514 can be
rendered into a single color-format image in a single render pass,
where each color information bit in the color-format image
corresponds to a respective visible object. This is useful as the
commonplace red green blue alpha (RGBA) texture has 32 color
information bits per pixel (8 bits per channel). Therefore, for
example, in one embodiment, 32 amodal segmentation images
associated with 32 visible objects are rendered into a single
32-bit (per pixel)-color-format image (e.g., an RGBA image) in a
single render pass. For example, at a particular pixel in the
32-bit-color-format image, if two objects whose amodal segmentation
image information occupies respectively the 6th and 7th least
significant bits in the color information are present, the
resulting color value for the pixel is 1100000 in binary (96 in
decimal), which is the result of bitwise OR of 1000000 in binary
(64 in decimal) and 100000 in binary (32 in decimal).
[0054] In one embodiment, with a multiple render target (MRT)
technique, even more amodal segmentation images can be renered in a
single render pass. In particular, a first plurality of amodal
segmentation images may be rendered into a second plurality of
color-format images in a single render pass, where each color
information bit in each of the color-format images corresponds to a
respective visible object. For example, in one embodiment, 128
amodal segmentation images associated with 128 visible objects are
rendered into four 32-bit (per pixel)-color-format images (e.g.,
RGBA images) in a single render pass.
[0055] Referring to FIG. 7, a flowchart illustrating an example
method 700 for generating tight two-dimensional (2D) bounding boxes
for visible objects in a three-dimensional (3D) scene according to
one embodiment is shown. The process 700 can be implemented in
hardware, software, or a combination thereof. At block 710, a
two-dimensional (2D) segmentation image of a three-dimensional (3D)
scene comprising one or more objects is generated by rendering the
3D scene with a segmentation camera. Each of the objects is
rendered in a single respective different color. At block 720, one
or more visible objects in the 3D scene are identified among the
one or more objects based on the segmentation image. At block 730,
a 2D amodal segmentation image for each of the visible objects in
the 3D scene is generated separately. Each amodal segmentation
image comprises only the single visible object for which it is
generated. Thereafter, at block 740, a 2D bounding box is generated
for each of the visible objects in the 3D scene based on the amodal
segmentation image for the visible object.
[0056] FIG. 8 shows an example of a perception training module
according to one embodiment. Perception training module 800 may be
implemented as part of perception training system 125 of FIG. 1 for
training a perception system of an autonomous driving system such
as perception module 302. Referring to FIG. 8, a block diagram
illustrating an example apparatus 800 according to one embodiment
is shown. Various modules illustrated in FIG. 8 can be implemented
in either hardware or software. A 2D segmentation image generation
module 802 is configured to generate a 2D segmentation image of a
3D scene comprising one or more objects by rendering the 3D scene
with a segmentation camera. Each of the objects is rendered in a
single respective different color. A visible object identifying
module 804 is configured to identify one or more visible objects in
the 3D scene among the one or more objects based on the
segmentation image. A 2D amodal segmentation image generation
module 806 is configured to generate a 2D amodal segmentation image
for each of the visible objects in the 3D scene separately. Each
amodal segmentation image comprises only the single visible object
for which it is generated. A 2D bounding box generation module 808
is configured to generate a 2D bounding box for each of the visible
objects in the 3D scene based on the amodal segmentation image for
the visible object.
[0057] Note that some or all of the components as shown and
described above may be implemented in software, hardware, or a
combination thereof. For example, such components can be
implemented as software installed and stored in a persistent
storage device, which can be loaded and executed in a memory by a
processor (not shown) to carry out the processes or operations
described throughout this application. Alternatively, such
components can be implemented as executable code programmed or
embedded into dedicated hardware such as an integrated circuit
(e.g., an application specific IC or ASIC), a digital signal
processor (DSP), or a field programmable gate array (FPGA), which
can be accessed via a corresponding driver and/or operating system
from an application. Furthermore, such components can be
implemented as specific hardware logic in a processor or processor
core as part of an instruction set accessible by a software
component via one or more specific instructions.
[0058] Some portions of the preceding detailed descriptions have
been presented in terms of algorithms and symbolic representations
of operations on data bits within a computer memory. These
algorithmic descriptions and representations are the ways used by
those skilled in the data processing arts to most effectively
convey the substance of their work to others skilled in the art. An
algorithm is here, and generally, conceived to be a self-consistent
sequence of operations leading to a desired result. The operations
are those requiring physical manipulations of physical
quantities.
[0059] It should be borne in mind, however, that all of these and
similar terms are to be associated with the appropriate physical
quantities and are merely convenient labels applied to these
quantities. Unless specifically stated otherwise as apparent from
the above discussion, it is appreciated that throughout the
description, discussions utilizing terms such as those set forth in
the claims below, refer to the action and processes of a computer
system, or similar electronic computing device, that manipulates
and transforms data represented as physical (electronic) quantities
within the computer system's registers and memories into other data
similarly represented as physical quantities within the computer
system memories or registers or other such information storage,
transmission or display devices.
[0060] Embodiments of the disclosure also relate to an apparatus
for performing the operations herein. Such a computer program is
stored in a non-transitory computer readable medium. A
machine-readable medium includes any mechanism for storing
information in a form readable by a machine (e.g., a computer). For
example, a machine-readable (e.g., computer-readable) medium
includes a machine (e.g., a computer) readable storage medium
(e.g., read only memory ("ROM"), random access memory ("RAM"),
magnetic disk storage media, optical storage media, flash memory
devices).
[0061] The processes or methods depicted in the preceding figures
may be performed by processing logic that comprises hardware (e.g.
circuitry, dedicated logic, etc.), software (e.g., embodied on a
non-transitory computer readable medium), or a combination of both.
Although the processes or methods are described above in terms of
some sequential operations, it should be appreciated that some of
the operations described may be performed in a different order.
Moreover, some operations may be performed in parallel rather than
sequentially.
[0062] Embodiments of the present disclosure are not described with
reference to any particular programming language. It will be
appreciated that a variety of programming languages may be used to
implement the teachings of embodiments of the disclosure as
described herein.
[0063] In the foregoing specification, embodiments of the
disclosure have been described with reference to specific exemplary
embodiments thereof. It will be evident that various modifications
may be made thereto without departing from the broader spirit and
scope of the disclosure as set forth in the following claims. The
specification and drawings are, accordingly, to be regarded in an
illustrative sense rather than a restrictive sense.
* * * * *